kern/132827: [vm] [panic] vm_page_free_toq: freeing mapped page

[Steven G. Kargl]

The following reply was made to PR kern/132827; it has been noted by GNATS.

From: "Steven G. Kargl" <kargl at troutmask.apl.washington.edu>
To: Alan Cox <alc at cs.rice.edu>
Cc: bug-followup at FreeBSD.org
Subject: Re: kern/132827: [vm] [panic] vm_page_free_toq: freeing
 mapped page
Date: Wed, 27 May 2009 11:43:46 -0700 (PDT)

 Alan Cox wrote:
 > Steven G. Kargl wrote:
 > > Alan Cox wrote:
 > >   
 > >> There has been no reported re-occurrence of this assertion failure since 
 > >> the submitter removed ZERO_COPY_SOCKETS from his kernel configuration.
 > >>     
 > >
 > > Yes, that is correct.  The panics stopped after removal of
 > > ZERO_COPY_SOCKETS.  I'd suggest adding a big warning in
 > > the various NOTES files that ZERO_COPY_SOCKETS is broken.
 > >
 > >   
 > 
 > I will gladly commit appropriately frightening language.  Send me a 
 > patch.  :-)
 > 
 
 Index: NOTES
 ===================================================================
 --- NOTES       (revision 191256)
 +++ NOTES       (working copy)
 @@ -930,7 +930,9 @@
  # receiving data via a socket.  The send side works for any type of NIC,
  # the receive side only works for NICs that support MTUs greater than the
  # page size of your architecture and that support header splitting.  See
 -# zero_copy(9) for more details.
 +# zero_copy(9) for more details.  This option is assocated with a rare,
 +# but possible, kernel panic in vm_page_free_toq().  See kern/132827 for
 +# additional details.
  options        ZERO_COPY_SOCKETS
 
 -- 
 Steve
 http://troutmask.apl.washington.edu/~kargl/